INDEX

Explanations

phrases related to rankings or positions within a competitive context

oai_token-act-pair · gpt-3.5-turbo

references to rankings or positions within a competitive context

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

apo

-0.75

ripp

-0.67

quit

-0.65

chn

-0.63

mb

-0.62

govtrack

-0.61

ching

-0.61

 Wond

-0.61

gone

-0.61

 Lauder

-0.60

POSITIVE LOGITS

 tier

0.97

iaries

0.94

ten

0.93

flight

0.92

 THREE

0.87

 spot

0.87

 eight

0.87

 drawer

0.87

 five

0.85

Activations Density 0.039%

phrases related to rankings or positions within a competitive context

references to rankings or positions within a competitive context

No Comments

No Known Activations